We assessed the short-term association between antidepressant drug use and the risk of new-onset diabetes in 2-years of observation. This study used longitudinal data from the Medical Expenditure Panel Survey for years 2004–2007. Chi-square tests and logistic regressions were used to examine the link between use of antidepressants with and without depression, and new-onset diabetes, after controlling for independent variables in blocks. In unadjusted models, the risk of new-onset diabetes was significantly increased for persons using antidepressants with depression compared with those without antidepressant use or depression [odds ratio (OR)=2.12, 95% confidence interval (CI): 1.45–3.09]. When lifestyle risk factors were entered in the model, statistical significance disappeared [adjusted OR (AOR)=1.42, 95% CI: 0.98–2.08]. Independently, lifestyle risk factors significantly increased the risk of new-onset diabetes: hypertension (AOR=2.55, 95% CI: 1.86–3.50, P<0.001), lipid disorders (AOR=1.60, 95% CI: 1.14–2.24), overweight (AOR=2.01, 95% CI: 1.35–2.98), obesity (AOR=3.57, 95% CI: 2.50–5.10), and no physical exercise (AOR=1.98, 95% CI: 1.53–2.57, P<0.001). Future studies on the risk of new-onset diabetes by duration and intensity of antidepressant use and depression are needed.